Detection of GAN-synthesized street videos

TL;DR

This paper introduces a simple frame-based detector for AI-generated street videos, demonstrating high accuracy even on compressed videos, addressing a gap in deepfake detection beyond facial videos.

Contribution

It presents the first effective detection method for AI-generated street videos, expanding deepfake detection to non-facial content with robust performance.

Findings

High detection accuracy on state-of-the-art DeepStreets videos

Robust performance on compressed videos with mismatched compression levels

Effective detection using a simple frame-based approach

Abstract

Research on the detection of AI-generated videos has focused almost exclusively on face videos, usually referred to as deepfakes. Manipulations like face swapping, face reenactment and expression manipulation have been the subject of an intense research with the development of a number of efficient tools to distinguish artificial videos from genuine ones. Much less attention has been paid to the detection of artificial non-facial videos. Yet, new tools for the generation of such kind of videos are being developed at a fast pace and will soon reach the quality level of deepfake videos. The goal of this paper is to investigate the detectability of a new kind of AI-generated videos framing driving street sequences (here referred to as DeepStreets videos), which, by their nature, can not be analysed with the same tools used for facial deepfakes. Specifically, we present a simple frame-based detector, achieving very good performance on state-of-the-art DeepStreets videos generated by the Vid2vid architecture. Noticeably, the detector retains very good performance on compressed videos, even when the compression level used during training does not match that used for the test videos.